The Oxford Study
Author
Discussion

Blue62

Original Poster:

10,324 posts

176 months

Wednesday 25th March 2020
quotequote all
Apologies for starting another thread but the main CV thread is moving so quickly that I thought this may be worth it’s own discussion.

Reported in the FT that there is a study, led by Prof Gupta that only 1:1000 will need treatment for the virus and that it may have been in the U.K. since mid Jan, infecting as much as half the population without us knowing it. Their modelling brings back into focus the herd immunity strategy, but it’s early stages and the programme of antibody testing is only just getting underway, with the involvement of Cambridge and Kent Uni’s.

Prof Gupta has been guarded in comments relating to the Imperial College response but has expressed surprise that it’s been so widely accepted.

cml24

1,558 posts

171 months

Wednesday 25th March 2020
quotequote all
I'm surprised this hasn't had as much airtime on the news.

It sounds like there are a few companies that are or are about to produce large quantities of antibody tests to check if people have previously had the virus. I didn't realise these were ready yet, but apparently the UK has bought 3.5million tests. That should give a good indication of how close to reality this latest study is.

anonymous-user

78 months

Wednesday 25th March 2020
quotequote all
I find this slightly bizarre though, Matt Hancock said 1 week ago that this test doesn’t exist yet

bazza white

3,730 posts

152 months

Wednesday 25th March 2020
quotequote all
To be fair things are moving at quite a swift pace.

Jasandjules

72,035 posts

253 months

Wednesday 25th March 2020
quotequote all
tux said:
I find this slightly bizarre though, Matt Hancock said 1 week ago that this test doesn’t exist yet
You trust him? Wow.

What I am most interested in is whether or not that cold/cough that so many had (me included) for a number of weeks over Christmas was Covid19 or not.

paulrockliffe

16,407 posts

251 months

Wednesday 25th March 2020
quotequote all
That study is based on a single assumption around the rate of hospitalisation. That rate might sit nicely in a model, but the real world evidence is that it's wrong. Take it with a pinch of salt.

Never forget; all models are wrong, a few are useful.

The underlying point is correct though; being able to test for immunity is critical to getting the economy back on track.

ATG

23,149 posts

296 months

Wednesday 25th March 2020
quotequote all
The only firm numbers you've got to work with are the number of people who have been tested. The earlier you assume the infection started spreading, the less serious you'll estimate the disease to be. I doubt enough people have been tested, particularly of those who show little or no sign of infection and have had no known exposure, to allow models to be calibrated with much confidence. If that data existed, there'd be no debate about when the disease entered the UK, the rate at which it spreads and how serious it is.

There is also no guarantee that those who have caught covid-19 will become immune to it in much the same way that we don't become immune to catching colds.

turbobloke

116,073 posts

284 months

Wednesday 25th March 2020
quotequote all
ATG said:
The only firm numbers you've got to work with are the number of people who have been tested. The earlier you assume the infection started spreading, the less serious you'll estimate the disease to be. I doubt enough people have been tested, particularly of those who show little or no sign of infection and have had no known exposure, to allow models to be calibrated with much confidence. If that data existed, there'd be no debate about when the disease entered the UK, the rate at which it spreads and how serious it is.

There is also no guarantee that those who have caught covid-19 will become immune to it in much the same way that we don't become immune to catching colds.
The signs have been good so far. The link below has free access if you haven't exceeded the free limit.

https://www.washingtonpost.com/health/the-coronavi...

Don Veloci

2,149 posts

305 months

Wednesday 25th March 2020
quotequote all
I had a viral infection in January that knocked me on my ass with fever and breathing trouble.

Highly unlikely it was Corona though. Given how contagious this thing seems to be I would have infected plenty at the time.



paulrockliffe

16,407 posts

251 months

Wednesday 25th March 2020
quotequote all
ATG said:
There is also no guarantee that those who have caught covid-19 will become immune to it in much the same way that we don't become immune to catching colds.
There's two elements to that; how quickly the virus mutates to something that we're not immune to and how long immunity lasts. The little stuff I've seen seems to suggest that the rate of mutation might be low, but the immunity isn't long lasting either. But there isn't any reliable knowledge on either at the moment.

The other thing we'll see in the next few months is whether the virus is impacted in the same way as the cold virus by UV light; the cold is seasonal because the transmission rate drops significantly when UV light levels rise. There is some evidence it does as cases are low where UV is high., but there are other reasons why that may be the case. Seasonality would be helpful.

sim72

4,998 posts

158 months

Wednesday 25th March 2020
quotequote all
It's an interesting idea. I presume that it posits that CV19 arrived here much earlier (possibly from Chinese students returning to universities around the world after Christmas), but is in fact less virulent that we have assumed, with most infected people being asymptomatic, and the few early deaths - being mostly the elderly and unwell - were written off as normal pneumonia victims. Extrapolating from that, it suggests that in fact the exponential curve didn't start at the first diagnosed death, but was already a significant way up that curve.

Blue62

Original Poster:

10,324 posts

176 months

Wednesday 25th March 2020
quotequote all
sim72 said:
It's an interesting idea. I presume that it posits that CV19 arrived here much earlier (possibly from Chinese students returning to universities around the world after Christmas), but is in fact less virulent that we have assumed, with most infected people being asymptomatic, and the few early deaths - being mostly the elderly and unwell - were written off as normal pneumonia victims. Extrapolating from that, it suggests that in fact the exponential curve didn't start at the first diagnosed death, but was already a significant way up that curve.
It's based on a susceptibility model based on case and death data from the UK and Italy, the researcher made a series of plausible assumptions based on what we know about the virus. Somebody made a point about mutation, my (limited) understanding is that in the vast majority of cases viruses become less virulent when they mutate, the only example I'm aware of where this s not the case is the Spanish Flu which killed more people in the second wave, though social and environmental conditions were a significant factor.

I am not saying it's a game changer, just good to see that there are other possibilities and it will be interesting to see what data emerges once antibody testing is underway. I think testing is the key to bringing this thing under control in any event, whether for infection or antibodies.



Hosenbugler

1,856 posts

126 months

Wednesday 25th March 2020
quotequote all
This is most interesting, I am , (one of many it seems) almost totally convinced that this virus has been here since the end of last year,or earlier, just not recognised as such.
My son is fit healthy , late 30's. A few days before he was to resume work during the Christmas/New year period, he woke with a raging fever , unable to get out of bed, that was where he spent the next 2 days . Again, he woke up in the morning, and it was like nothing had happened (albeit losing 8ib in weight) , whatever it was he had, it went as abruptly as it arrived. His O/H is a NHS staff nurse , and she was concerned, the fever seeming so aggressive , she now is of the opinion that what he had, was very likely Corona virus, he showed other symptons as well.

The potential of these tests to see if people have already had the virus, could be an eye opener, after all, retrospective tissue tests showed that aids had been around for many years before it was identified.

Meanwhile of course, we just have to follow what the CMO and his ilk recomend, and just hope this is all over sooner than later. Fingers crossed.

BlackTails

2,895 posts

79 months

Wednesday 25th March 2020
quotequote all
It's an interesting idea, but I'm a bit sceptical. I went down with a hideous bug on Boxing Day. Bit it was a regular nasty cold/flu bug in my eyes, in that it started in my throat/soft palate, went to work in my sinuses, gave me a runny nose and sneezing, blocked my nasal passages, then shifted to the disgusting thick green snot that I was snorting or sneezing out in volume, before rounding off with a brief stay in my chest. All with a high temp and a couple of days nailed to the sofa thrown in.

The thing I have had over the last few days was preceded with a sniffle no worse than you get walking outdoors on a cold day, a mild cough, then went straight and very rapidly (a couple of hours max) to a high temp plus fever, muscle aches, fatigues and coughing, necessitating bed rest and not just lounging on the sofa. I have had a sinus blockage and a sinus headache but there has been no head cold/snot phase to this at all (I don't remember having to use a single tissue), and I am now more or less through it.

Similar in some respects, but different in some important ones I think.

BoRED S2upid

20,996 posts

264 months

Wednesday 25th March 2020
quotequote all
Aren’t these tests also a bit dangerous? Say 3mill are tested 1 mill test positive but have shown no signs or very limited cough symptoms and have fully recovered, you release that data to the public whilst still trying to maintain the lockdown it’s just not going to work.

Oakey

27,970 posts

240 months

Wednesday 25th March 2020
quotequote all
Jasandjules said:
You trust him? Wow.

What I am most interested in is whether or not that cold/cough that so many had (me included) for a number of weeks over Christmas was Covid19 or not.
What were your symptoms?

Mid Jan I developed a really bad sore throat along with a hacking, dry cough and shortness of breath and a feeling like i was drowning. Unlike every other cold I'd had there was no runny / blocked nose which was really odd and it didn't feel as bad as any flu I'd had previously. This lasted for thee weeks and at the end I had fever like symptoms. Like Hosenbugler said, I also woke up a few days later and everything had simply gone. Ever since then I've had an odd pain in my chest, especially noticeable when inhaling.


Edited by Oakey on Wednesday 25th March 11:53

Blue62

Original Poster:

10,324 posts

176 months

Wednesday 25th March 2020
quotequote all
BoRED S2upid said:
Aren’t these tests also a bit dangerous? Say 3mill are tested 1 mill test positive but have shown no signs or very limited cough symptoms and have fully recovered, you release that data to the public whilst still trying to maintain the lockdown it’s just not going to work.
There's no suggestion of that whatsoever, the Oxford study supports current measures as they protect the ability go the NHS to deal with vulnerable people. It is exploring the possibility that the virus has been in our community for longer than we realise and that, in the majority of cases the symptoms are pretty mild. Current testing methods are only looking at rates of infection, this study will be looking at rates of antibodies, the implications if they are proved to be correct, are considerable on how we deal with CV.

DeltonaS

3,707 posts

162 months

Wednesday 25th March 2020
quotequote all
Dutch researchers in cooperation with bloodbanks are doing a similar test since last week. They're testing blood of regular blood donors, about 2.000 a day, on anti bodies

Obviously blood from blood donors normally go to a series of tests, they've added this to that list.

With enough data available they can estimate/extrapolate the number of people who have the virus and the development overtime, which will give a good picture of herd immunity against the coronavirus.

voyds9

8,490 posts

307 months

Wednesday 25th March 2020
quotequote all
Hosenbugler said:
This is most interesting, I am , (one of many it seems) almost totally convinced that this virus has been here since the end of last year,or earlier, just not recognised as such.
My son is fit healthy , late 30's. A few days before he was to resume work during the Christmas/New year period, he woke with a raging fever , unable to get out of bed, that was where he spent the next 2 days . Again, he woke up in the morning, and it was like nothing had happened (albeit losing 8ib in weight) , whatever it was he had, it went as abruptly as it arrived. His O/H is a NHS staff nurse , and she was concerned, the fever seeming so aggressive , she now is of the opinion that what he had, was very likely Corona virus, he showed other symptons as well.

The potential of these tests to see if people have already had the virus, could be an eye opener, after all, retrospective tissue tests showed that aids had been around for many years before it was identified.

Meanwhile of course, we just have to follow what the CMO and his ilk recomend, and just hope this is all over sooner than later. Fingers crossed.
Very similar to me, I never get ill.

After Christmas break I had 1 day off colour, 3 days of sleep couldn't keep my eyes open. Then back to work no after effects.
I have had quite a few people tell me similar tales from late December through early February.

Dr Z

3,396 posts

195 months

Wednesday 25th March 2020
quotequote all
https://twitter.com/SunetraGupta/status/1242403091...

^See the replies to this status for some friendly peer-review.